基于改进GWO和SVM的大坝变形预测

建立准确可靠的大坝变形预测模型是大坝安全评价的重要内容,为此,将差分进化算法的交叉和变异算子引入灰狼优化算法(GWO),提出一种基于改进灰狼算法(MGWO)优化支持向量机(SVM)的大坝变形预测方法。通过差分进化算法丰富初始种群,提出改进灰狼优化算法(MGWO),并采用MGWO算法优化SVM的惩罚因子和核函数,建立基于MGWO-SVM算法的大坝变形预测模型。以锦屏一级特高拱坝实测数据为例,将MGWO-SVM模型与SVM、GWO-SVM模型的预测结果进行比较。结果表明,MGWO-SVM模型可以有效提高大坝变形预测精度。

Dam Deformation Prediction Based on Modified Grey Wolf Optimization Algorithm and Support Vector Machine

The establishing of an accurate and reliable dam deformation prediction model is an important content of dam safety evaluation.To this end,a dam deformation prediction method based on support vector machine(SVM)with optimal parameters selected by modified grey wolf optimization algorithm(MGWO)is proposed by introducing the crossover and mutation operators of the differential evolution algorithm into the GWO algorithm.The initial population is enriched by the differential evolution algorithm to propose an MGWO and the penalty factor and kernel function of SVM are optimized by the MGWO algorithm,then a dam deformation prediction model based on the MGWO-SVM algorithm is established.The results are compared with those of the SVM and GWO-SVM models through the measured data of Jinping I super high arch dam.The research shows that the proposed MGWO-SVM model can effectively improve the accuracy of dam deformation prediction.